Data Augmentation Based on Attributed Scattering Centers to Train Robust CNN for SAR ATR
Open Access
- 25 February 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 25459-25473
- https://doi.org/10.1109/access.2019.2900522
Abstract
Driven by the good classification performance of convolutional neural network (CNN), this study proposes a CNN-based synthetic aperture radar (SAR) target recognition method. Considering the scarce training samples in the field of SAR target recognition, a novel data augmentation algorithm is designed through target reconstruction based on attributed scattering centers (ASC). ASCs reflect the electromagnetic phenomenon of SAR targets, which can be used to reconstruct the whole or part of the target. The sparse representation (SR) algorithm is first employed to extract the ASCs from a single SAR image. Afterwards, a subset of the extracted ASCs are selected to reconstruct the target’s image. By repeating the process, many new images can be generated as available training samples. In the classification stage, a CNN architecture is designed and trained by the augmented samples. As for the test sample, it is also reconstructed using all its extracted ASCs thus relieving the interferences caused by the clutters or noises in the background. Finally, the reconstructed image from the test sample is classified based on the trained CNN. The reconstructed image from ASCs can reduce the clutters and noises thus enhancing the image quality. More importantly, the generated new training samples could cover more operating conditions, which may probably occur in SAR target recognition. Therefore, the trained CNN can work more robustly under different situations. In the experiments, the moving and stationary target acquisition and recognition (MSTAR) dataset is used to evaluate the performance of the proposed approach. This method could classify the 10 classes of targets with an accuracy of 99.48% under the standard operating condition (EOC). For the extended operating conditions like configuration variance, depression angle variance, noise corruption, and partial occlusion, the proposed method also displays superior performance over some baseline algorithms drawn from state-of-the-art literatures.Keywords
This publication has 41 references indexed in Scilit:
- Neighborhood Geometric Center Scaling Embedding for SAR ATRIEEE Transactions on Aerospace and Electronic Systems, 2014
- Automatic target recognition of synthetic aperture radar (SAR) images based on optimal selection of Zernike moments featuresIET Computer Vision, 2014
- SVM-based target recognition from synthetic aperture radar images using target region outline descriptorsNonlinear Analysis, 2009
- Robust automatic target recognition using learning classifier systemsInformation Fusion, 2007
- Support vector machines for SAR automatic target recognitionIEEE Transactions on Aerospace and Electronic Systems, 2001
- Performance of 10- and 20-target MSE classifiersIEEE Transactions on Aerospace and Electronic Systems, 2000
- Model-based classification of radar imagesIEEE Transactions on Information Theory, 2000
- Recognition of articulated and occluded objectsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1999
- Attributed scattering centers for SAR ATRIEEE Transactions on Image Processing, 1997
- Approximation by superpositions of a sigmoidal functionMathematics of Control, Signals, and Systems, 1989